Flying Is Safer Than Driving — Depending Entirely on How You Count

The Safest Way to Travel, Says Everyone

Flying is the safest form of transport. Airlines say it, safety regulators say it, and every nervous flyer has heard it from a friend while gripping an armrest. The supporting numbers are impressive. IATA’s 2024 safety report counted 40.6 million commercial flights, seven fatal accidents and 244 deaths on board. Their calculation is that a person would have to fly every single day for about 15,000 years before expecting to die in a crash. In 2023 the industry managed a single fatal accident worldwide.

Against that, roads killed around 41,000 people in the United States alone in 2023, and roughly 1.2 million worldwide. On these figures, flying is not merely safer than driving. It is in a different category.

The claim is true. What it hides is that the size of the gap depends on the unit you measure it in, and that none of the usual units say anything about the number that decides your own risk, which is how often you fly.

Per Kilometre, Per Hour, Per Trip

The “safest form of transport” comparison is almost always made per passenger-kilometre. Deaths divided by distance travelled. This is the unit that makes flying look best, because a flight covers enormous distances in a single, mostly uneventful cruise. Divide a handful of deaths by trillions of kilometres and the rate collapses toward zero. Cars, which rack up their kilometres in short trips through intersections, look terrible by the same measure. On US figures, driving comes out at roughly 8 deaths per billion kilometres against about 0.03 for scheduled airlines, a gap of two hundred to one.

Per hour tells a different story. An average flight lasts two to three hours, and nearly all the risk is in the takeoff and landing, so the per-hour rate does not benefit from the cruise the way the per-kilometre rate does. Using the same underlying data, driving comes out at about ten times riskier than flying per hour. Still safer, but the gap has shrunk twentyfold.

Per trip is where people usually go wrong, in both directions. An average car trip in the US is about 15 kilometres, which at 8 deaths per billion kilometres works out to roughly one in eight million. A commercial flight carries a per-boarding fatality risk of about one in ten million. Put side by side, those look like a tie, and that comparison shows up in plenty of “actually, flying is no safer” arguments. It is a bad comparison. Nobody decides between a run to the supermarket and a flight to Denver. The real choice is how to cover one specific distance. Drive 800 kilometres, Chicago to Atlanta say, and the trip’s fatality risk is about one in 150,000. Fly it, and it stays at one in ten million. For any journey where flying is a genuine alternative, boarding the plane is sixty to a hundred times safer per decision.

So the three units agree on the direction and disagree on the size. Per kilometre, two hundred to one. Per hour, ten to one. Per substitutable journey, somewhere around sixty to one. All of them answer the question “which vehicle should I take for this trip”. None of them answer the question “how much risk does flying add to my year”, because that depends on a number the comparison never mentions: how many trips you actually take.

The Exposure Nobody Counts

An average American makes something like a thousand car trips a year and boards a commercial flight about twice. Multiply each by its per-trip risk and the year’s driving risk comes out several hundred times the flying risk. Part of that gap is the vehicle: a car shares the road with thousands of other human drivers, a jet flies in monitored, separated airspace with two rested pilots. But most of it is simple frequency. The car gets used five hundred times more often.

That reframing changes who the “flying is safe” message applies to. For an occasional flyer, the annual risk from flying is negligible next to the drive to the airport. For a consultant on 150 segments a year, a pilot, or cabin crew, the flying risk starts to look like a real line item. Not a large one, but not a rounding error either.

There is also the matter of what “flying” means. The IATA numbers cover scheduled commercial airlines. Small private aircraft in the United States run a fatal accident rate around one per 100,000 flight hours, twenty to forty times the airline rate. Helicopter tours, bush flying, and charter operations in parts of the world with weaker oversight sit somewhere in between. “Flying is safe” is a statement about a Boeing or Airbus operated by a major carrier, and it does not travel well.

The Chain of Probabilities

The calculator uses two fixed steps. First, the chance that a given flight becomes a fatal accident: on the 2020-2024 average this is roughly one in five million departures. 2024 was a worse year than 2023, with seven fatal accidents against one, and the five-year average smooths that out.

Second, the chance that you personally die given that your flight has a fatal accident. That sounds like it should be one, and it isn’t. “Fatal accident” means at least one death on board. Runway excursions, hard landings and post-crash fires often kill some passengers and not others. IATA’s fatality-risk figure, which counts deaths per passenger rather than accidents per flight, comes out at about half the fatal-accident rate. So the calculator uses a one-in-two chance of being among the dead.

Together those give a per-flight fatality risk of roughly one in ten million. The only slider is how many flights you board.

Your Actual Risk

At four flights a year, a return trip with one connection each way, the annual risk is about 1-in-2.5 million. That is a thousandth of the annual risk of dying in a car, and well below lightning. At 50 flights a year, a heavy business-travel schedule, it reaches 1-in-200,000. At the slider’s maximum of 300 flights, roughly what a short-haul pilot or cabin crew member flies, it passes 1-in-33,000, which is still a quarter of the annual driving risk.

That last comparison is the honest version of the airline claim. Even the people who fly for a living face less risk from flying than an ordinary commuter faces from driving. The message is true. It just isn’t true because of the per-kilometre number that usually gets quoted.

Sources

• IATA (2025). 2024 Annual Safety Report. 40.6 million flights, 7 fatal accidents, 244 on-board fatalities. Fatality risk 0.06 per million sectors in 2024; five-year (2020-2024) average 0.10.
• IATA (2024). 2023 Annual Safety Report. One fatal accident (Yeti Airlines flight 691, Nepal, 72 deaths) in about 37 million flights.
• NHTSA (2024). Early estimate of motor vehicle traffic fatalities in 2023: about 40,990 deaths, roughly 1.26 per 100 million vehicle miles travelled.
• WHO (2023). Global Status Report on Road Safety. Approximately 1.19 million road deaths per year worldwide.
• US Federal Highway Administration, National Household Travel Survey. Average car trip length about 9 to 10 miles.
• NTSB and FAA general aviation statistics. Fatal accident rate for US general aviation roughly 1 per 100,000 flight hours.